Fang Ren, Ph.D.
Douglas M. Flewelling, Ph.D.
The automotive repossession industry has been transitioning to locationally aware hardware that permits the use of geographic data to enhance business intelligence operations. GIS initiatives in this industry are relatively new meaning that geospatial trends in repossession data have not been studied in an academic context. This project focuses on identifying trends in license plate recognition scan data collected from high-resolution cameras in Houston, Texas in order to predict future repossession locations. Through the use of opportunity terrain modeling with spatial statistics, a prediction surface was generated that accurately described the habitat of debt by combining seven opportunity variables that were significantly correlated with repossession densities. The findings can be used to narrow the search for vehicles by targeting high-opportunity locations for scanning and recovery.
Sachdev, V. (2015). Refining the Search and Recovery Process: A Predictive Model for Vehicle Repossessions (Master's thesis, University of Redlands). Retrieved from http://inspire.redlands.edu/gis_gradproj/242